Ricardo Nazaré
Serrazina
Cermetos à base de Zircónia sintetizada
por Detonação de Emulsão (EDS)
Universidade de Aveiro 2016
Department of Materials and Ceramic Engineering
Emulsion Detonation Synthesis (EDS)
Zirconia-based CERMETs
Ricardo Nazaré
Serrazina
Cermetos à base de Zircónia sintetizada
por Detonação de Emulsão (EDS)
Emulsion Detonation Synthesis (EDS)
Zirconia-based CERMETs
Ricardo Nazaré
Serrazina
Cermetos à base de Zircónia sintetizada
por Detonação de Emulsão (EDS)
Internship report presented to the University of Aveiro in the fulfillment of the requirement for the awarding of the Masters in Materials Science and Engineering carried out under the supervision of Prof. Doutora Ana Maria Oliveira e Rocha Senos, Associate Professor in the Department of Materials and Ceramics Engineering, co-supervision of Prof. Doutora Paula Maria Lousada Silveirinha Vilarinho, Associate Professor in the Department of Materials and Ceramics Engineering and internship supervision of Doutor Nuno Miguel Pinto Neves and Drª Rosa Filomena Duarte Vitória Calinas.
Important note:
Due to confidentiality issues, parts of this internship report are deleted or covered with a “Confidential” stamp.
Emulsion Detonation Synthesis (EDS)
Zirconia-based CERMETs
The Board of Examiners
President Prof. Doutor Joaquim Manuel Vieira
Full professor from University of Aveiro, Portugal
Prof. Doutor Albano Augusto Cavaleiro Rodrigues de Carvalho
Full professor from University of Coimbra, Portugal
Prof. Doutora Ana Maria de Oliveira e Rocha Senos
Acknowledgements
First of all, I would like to thank my supervisors, both from University and from INNOVNANO, for all orientation and support during the realization of this work.I would like to thank the University of Aveiro for giving me the conditions to perform this work, and especially to all DEMaC’s STAFF for all help and support. I would like to personally thank to Eng. Célia Miranda, Eng. Artur Sarabando, Engª. Marta Ferro and Eng. Tiago Silva.
To my lab coworkers I am grateful for the experimental support and orientation. I would like to highlight the help of Dr. Sebastian Zlotnik, Monica Tomczyk, Pedro Lemos Marques and Dra. Cristina Fernandes.
To Dr. Venkata Ramana that was an essential help to perform and interpret magnetic and Raman studies. I am extremely grateful to my parents to allow me to be able of having a master degree and to support me in all of my decisions.
To Cátia Ferreira and Cristina Mesquita from INNOVNANO, I want to thank to all the experimental help and guidance. I would like to thank to my girlfriend that always supported, listened and helped me during the good and bad moments. To all my master’s colleagues I am also grateful for the supporting and advising.
To everyone who directly and indirectly contributed to this work, I am grateful for support in the fulfillment of my dream of working in science research and at such great organization as DEMaC/University of Aveiro.
keywords
CERMETs, Ceramic matrix composites, Yttria stabilized Zirconia, Metal alloy, EDS, Mechanosynthesis.summary
This work main goal is to develop metal composites on a ceramic matrix, designated as CERMETs. Hence, and having as starting materials the ceramic and composite powders produced by INNOVNANO, mechanosynthesis powders were also prepared. The yttria stabilized zirconia (YSZ) based powders (produced by those two methods) were crystallographic, chemically, morphologically, rheological, thermal and magnetically characterized. Pressed compacts from the previous powders were prepared and sintered in vacuum conditions, followed by their characterization namely in terms of structure, microstructure, mechanical and magnetic behavior.The results were analyzed with a critical mind, trying to co-relate the physical and chemical properties of the powders and green compacts with the final sintered properties. The comparison between ceramic and composites (either powders, either sintered compacts) was always the main goal during the development of this work.
The results include the evaluation of the meaning and importance of the several powders preparation steps that are conducted in INNOVNANO’s producing method, the importance of the tetragonal zirconia phase stabilization, the magnetic response and the interpretation of the decrease in mechanical resistance verified in CERMETs.
palavras-chave CERMETOs, Compósitos de matriz cerâmica, Zircónia estabilizada com Ítria, liga metálica, EDS, Mecanossíntese.
resumo
A realização deste trabalho tem como objetivo a produção e caracterização de compósitos de metal em matriz cerâmica (CERMETOS). Assim sendo, e tendo como base pós cerâmicos e compósitos produzidos pela INNOVNANO, foram também produzidos pós por mecanossíntese.Os pós à base de zircónia estabilizada com ítria (YSZ), produzidos por estes dois métodos, foram caracterizados cristalográfica, química, morfológica, reológica, térmica e magneticamente. Os pós foram compactados e sinterizados em vácuo, seguindo-se a sua caracterização, principalmente estrutural, microestrutural, mecânica e magnética.
Todos os resultados foram analisados com olhar crítico e tentando relacionar as propriedades físicas e químicas dos pós e compactos verdes com as propriedades finais dos sinterizados. A comparação entre cerâmicos e compósitos (quer pós, quer sinterizados) foi sempre o principal objetivo deste trabalho. Os resultados incluem a avaliação da importância das várias etapas de preparação dos pós, essencialmente no processo produtivo da INNOVNANO, a avaliação da estabilização da fase tetragonal da zircónia, a resposta magnética e a interpretação do decréscimo nas propriedades mecânicas verificado nos compósitos.
i Figure Index ... v Table index ... xi Abbreviations ... xiii Symbols ... xv 1. Introduction... 1 1.1 CERMETs ... 3
1.2 Yttria-stabilized Zirconia (YSZ) ... 5
1.2.1. Phase characterization and transformation toughening ... 5
1.2.2. Phase diagrams ... 8
1.2.3. YSZ Innovnano products ... 11
1.2.3.1. Innovnano, the Company ... 11
1.2.3.2. EDS technology ... 11
1.2.3.3. 2YSZ ... 13
1.2.3.4. 3YSZ ... 14
1.3. Interaction between metal and ceramic matrix ... 15
1.4. TiC CERMETs ... 17
1.5. WC CERMETs... 18
1.5.1. WC matrix with (Co/Ni) ... 18
1.5.2. WC-Co matrix with YSZ ... 20
1.5.3. Ni/W matrix with ZrO2 ... 21
1.6. Zirconia composites ... 22
1.6.1. Zirconia matrix with TiO2 and Fe3O4 ... 22
1.6.2. Ordering in Zirconia/Nickel composites ... 23
1.6.3. YSZ composites ... 25
1.6.3.1. YSZ matrix with WC composites ... 25
ii
1.6.4.3. Enhanced mechanical properties ... 30
1.6.4.4. Magnetic behavior ... 33
1.6.4.5. Sintering studies of YSZ+Ni CERMETs ... 36
1.7. Motivation and main goals ... 40
2. Experimental procedure ... 41
2.1 Materials ... 43
2.1.1. Powder production ... 44
2.1.1.1. EDS powders ... 44
2.1.1.2. Mechanosynthesized powders ... 44
2.1.2. Pressing and calcination ... 47
2.1.3. Sintering ... 48
2.2 Characterization techniques ... 50
2.2.1. Crystallographic characterization... 50
2.2.1.1. XRD: X-Ray diffraction ... 50
2.2.1.2. Raman spectroscopy ... 51
2.2.2. Morphologic and chemical characterization ... 52
2.2.2.1. SEM ... 52
2.2.2.2. TEM ... 54
2.2.2.3. ICPS... 55
2.2.2.4. XRF: X-ray fluorescence ... 56
2.2.3. Particle and grain size analysis ... 57
2.2.3.1. CPS... 57
2.2.3.2. Malvern Laser diffraction ... 58
2.2.3.3. BET ... 58
2.2.3.4. Sintered pellets grain size ... 60
iii 2.2.4.3. Archimedes method ... 63 2.2.5. Thermal analysis ... 63 2.2.5.1. DTA/TG ... 63 2.2.5.2. Dilatometry ... 64 2.2.6. Electrical conductivity... 64 2.2.7. Thermal conductivity ... 65 2.2.8. Mechanical characterization ... 66 2.2.8.1. Flexural strength ... 66 2.2.8.2. Hardness ... 66 2.2.9. Magnetic behavior ... 68
3. Results and discussion ... 71
3.1 Powders characterization ... 73
3.1.1. YSZ powders ... 73
3.1.2. Metal alloy, M ... 83
3.1.3. CERMET powders produced by EDS ... 85
3.1.4. CERMET powders produced by mechanosynthesis ... 93
3.2. Sintered products characterization ... 100
3.2.1. Structural and Microstructural characterization ... 100
3.2.1.1. Ceramic parts ... 100
3.2.1.2. CERMET parts... 104
3.2.2. Mechanical characterization ... 115
3.2.3. Thermal and electrical conductivity ... 119
3.3. Magnetic characterization ... 121
4. Conclusion ... 125
v
Figure 1.1 – Representation of crystallographic structures of Zirconia 16. ... 6
Figure 1.2 – Phase diagram of Zirconia-Yttria system. The figure shows the zirconia rich zone (mol% of ZrO2 > mol% of Y2O3) 15. ... 9
Figure 1.3 - Partial phase diagram of bulk YSZ, with indication of T0 and T’0 lines, being T0 the M/T
T-zero temperature line and T’0 is the T/C T-zero temperature line. Adapted from 19. ... 10
Figure 1.4 – Optical micrographs of a composite prepared from WC powders coated with iron rich binders (a) and a conventionally prepared WC composite with the same binders (b). The surrounding area shows the heterogeneity of binder distribution. 38. ... 19
Figure 1.5 – Effects of gradual cobalt substitution by nickel on magnetic saturation (CoM) and coercive force or coercivity (Hc) of WC/6Co alloy 37. ... 20
Figure 1.6 – Typical hysteresis loop of solid superacid of calcined tetragonal stabilized zirconia 43.
... 23 Figure 1.7 – SEM micrographs of monolithic composites of mullite/molybdenum with 34 vol.% of Mo (A) and ZrO2/Ni with 18 vol.% of Ni (B). At the bottom, theoretical models representing
different topologies: disordered (C) and ordered (D) particles. The mean first-neighbor distance <r> is well defined only in (D) 7,45. ... 24
Figure 1.8 – Microstructure of a ZrO2-based composite with 40 vol% WC, showing the existence
of bright WC nanoparticles in a dark ZrO2 matrix 18. ... 26
Figure 1.9 – HRTEM micrographs of 3Y-TZP/Ni interfaces found in nancomposites (A) and microcomposites (B). In (A), from the angular orientation of (111) Ni planes and (112) Zirconia planes with respect to the interface, it has been assigned as ZrO2(002)/Ni(110) 7. ... 29
Figure 1.10 – TEM picture of a 5 vol.% Ni in 3Y-TZP. (a) small particles covered with amorphous layer; (b) single crystals of Ni; (c) large Ni particles – grain size similar to YSZ one. Adapted from
6. ... 30
Figure 1.11 – Vickers hardness (HV) of 3Y-TZP/Ni nanocomposites. The blank squares represent two samples of micrometer composites (Ni content with d50 = 2µm). The dashed lines represent
theoretical hardness calculated assuming the linear rule of mixtures (considering 10.8 and 0.6 GPa for YSZ and Ni respectively) 6... 31
Figure 1.12 – Magnetization curves of Y-TZP/0.3 mol% NiO solid solution and Y-TZP/Ni nanocomposite after the internal reduction treatment measured by the super-conducting quantum interference magnetometer at room temperature 47. ... 35
vi
9 s was amplified in the top-left corner. Right: Temperature dependence of M-H loops for the
smaller particle size sample 9. ... 36
Figure 1.14 – (a) Effect of sintering temperature on the phase composition of YSZ+2 wt.% Ni. M, c and t refer to Monoclinic, Cubic and Tetragonal phase, respectively. Bottom XRD spectrum is before sintering. ... 38
Figure 1.15 – Effect of NiO on the microstructure of YSZ sintered at 1400°C: (a) YSZ; (b) 0.5 wt.% Ni; (c) 0.5 wt.% NiO; and (d) 2 wt.% Ni 46. The grain size is clearly increasing with the amount of NiO. ... 39
Figure 2.1 – Powders production scheme. ... 43
Figure 2.2 – Scheme of the experimental procedure for powder processing and characterization. ... 44
Figure 2.3 – A: High speed ball milling apparatus. B: Schematic representation of the movement of the material and media inside the bowl in a planetary high speed ball milling 64. C: Scheme of mechanosynthesis process. ... 46
Figure 2.4 – Scheme of the complete process since powders until sintering pellets obtaining. 48 Figure 2.5 – Picture of the furnace, temperature register and vacuum system. ... 49
Figure 2.6 – A: Sintering cycle schematic representation. B: Alumina crucibles used in the sintering process. ... 49
Figure 2.7 – X-Ray operation scheme 70. ... 50
Figure 2.8 – Schematic representation of the sample preparation for SEM. To see more about “Polishing and gridding in INNOVNANO” see Table 2.4. ... 54
Figure 2.9 – SEM Hitachi S-4100 (A) and STEM Hitachi SU-70 (B) (DEMaC)... 54
Figure 2.10 – TEM Hitachi H9000 equipment (DEMaC). ... 55
Figure 2.11 – Typical hysteresis loop of a magnetic sample. ... 69
Figure 3.1 – XRD patterns for YSZ powders: 2Y (DET.), 2Y and 2Y (PA). The crystallographic phases that were identified are recognized with symbols: • for tetragonal ZrO2 and for monoclinic ZrO2. (A) and (B) show the same data but with linear and logarithmic (base 10) scale bar in OY axis, respectively. ... 74
Figure 3.2 – Raman spectra of YSZ powders (2Y (DET.) and 2Y). The tetragonal zirconia peaks are identified with wavelength values in red and the monoclinic zirconia ones with yellow. ... 75
Figure 3.3 – SEM micrographs of YSZ powders: 2Y (DET.), 2Y and 2Y (PA). ... 77
Figure 3.4 – TEM micrographs of 2Y (DET.) powder. ... 77
vii
and respective derivative curves (with dots). The analysis was performed with 10 °C/min heating
rate. ... 81
Figure 3.7 – XRD pattern for the metal alloy M under study in this work. The crystallographic phase that was identified was only the metal alloy. OY axis bar is in linear scale. ... 83
Figure 3.8 – SEM micrographs of the metal alloy. ... 84
Figure 3.9 – SEM-EDS of the metal alloy. ... 84
Figure 3.10 – TG and DTA of the metal alloy in air. ... 84
Figure 3.11 - XRD patterns for EDS-CERMET powders: 2Y+M (DET.) and 2Y+M (PA). The crystallographic phases that were identified are recognized with symbols: • for tetragonal ZrO2, for monoclinic ZrO2 and * for the metal alloy M. (A) and (B) show the same data but with linear and logarithmic (base 10) scale bar in OY axis, respectively. ... 85
Figure 3.12 - Raman spectra of EDS-CERMET powders (2Y+M (DET.) and 2Y+M (PA)). The ceramic powder 2Y is represented for comparison. The tetragonal zirconia peaks are identified with the values in red and the monoclinic zirconia ones with yellow. ... 86
Figure 3.13 - SEM micrographs of EDS-CERMET powders: 2Y+M (DET.) and 2Y+M (PA). ... 87
Figure 3.14 - SEM-EDS of EDS-CERMET powders: 2Y+M (DET.) and 2Y+M (PA)... 88
Figure 3.15 - SEM micrographs of ceramic (2Y and 2Y (PA)), CERMET (2Y+M (PA)) and metal alloy (M) powders, impregnated in araldite and polished. ... 88
Figure 3.16 – TEM micrographs of EDS-CERMET powders: 2Y+M (DET.) and 2Y+M (PA). ... 89
Figure 3.17 - Aggregate/agglomerate size distribution of EDS-CERMET powders determined by Malvern. ... 90
Figure 3.18 – Comparison of dilatometric curves in argon and air for: (A): 2Y+M (DET.) and (B) 2Y+M (PA) EDS-CERMET compacts. In bold, the dilatometric curves and respective derivatives, with dots. Heating rate = 10 °C/min. ... 91
Figure 3.19 - EDS-CERMET powders, 2Y+M (DET.) 2Y+M (PA), dilatometric curves (in bold) and respective derivatives (with dots) in argon. 2Y (PA) sample dilatometry and derivative curve is also shown for comparison purposes. Heating rate = 10 °C/min. ... 92
Figure 3.20 - XRD patterns for MS-CERMET powders: MS-1M, MS-2M and MS-Y2M. The identified crystallographic phases are marked with symbols: • for tetragonal ZrO2 and for monoclinic ZrO2 and * for the metallic phase. (A) and (B) show the same data but with normal and logarithmic scale bar in OY axis, respectively. ... 93
Figure 3.21 - Raman spectra of MS-CERMET powders. The tetragonal zirconia peaks are identified with the values in red and the monoclinic zirconia ones with yellow. ... 94
viii
Figure 3.24 – TEM micrographs of MS-CERMET powders: MS-1M, MS-2M. ... 96 Figure 3.25 - Aggregate size distribution of MS-CERMET powders determined by Malvern ... 97 Figure 3.26 – Dilatometric analysis in argon and air (at bold), and respective derivative curve (with dots) s of MS-CERMET powders; (A): MS-1M, (B): MS-2M, and (C): MS-Y2M. Heating rate = 10 °C/min. ... 98 Figure 3.27 – A: MS-CERMET powders, MS-1M, MS-2M and MS-Y2M, dilatometric curves (bold) and respective derivatives (with dots) in argon. B: Dilatometric analysis of EDS CERMETs already presented, here for comparison effects only. Heating rate = 10 °C/min. ... 99 Figure 3.28 – XRD patterns of sintered ceramics (2Y and 2Y (PA)). ... 101 Figure 3.29 - SEM micrographs of sintered ceramics: 2Y and 2Y (PA). ... 103 Figure 3.30 – Sintered ceramic samples grain size distribution (nm) in histograms. ... 103 Figure 3.31 – XRD patterns of sintered EDS-CERMET (2Y+M (PA)) and CERMETs (1M, MS-2M and MS-YMS-2M). ... 104 Figure 3.32 - Raman spectra of sintered ceramic (2Y) and sintered CERMETs (2Y+M (PA), MS-1M and MS-2M). The tetragonal zirconia peaks are identified with the values in red and the monoclinic zirconia ones with blue... 106 Figure 3.33 - SEM micrographs of sintered CERMETs: 2Y+M (PA), MS-1M, MS-2M and MS-Y2M. ... 109 Figure 3.34 – EDS spectra of some CERMET samples, performed on the red circles areas of Figure 3.33. ... 110 Figure 3.35 - Sintered CERMET samples grain size distribution (nm) in histograms. ... 110 Figure 3.36 – SEM-EDS map of 2Y (PA) sample (B) and respective micrograph that represents the analyzed area (A). Magnification: 3000x. ... 111 Figure 3.37 – SEM-EDS map of 2Y+M (PA) sample (B) and respective micrograph that represents the analyzed area (A). Magnification: 3000x. ... 112 Figure 3.38 – SEM-EDS map of MS-1M sample (B) and respective micrograph that represents the analyzed area (A). Magnification: 3000x. ... 112 Figure 3.39 – SEM-EDS map of MS-2M sample (B) and respective micrograph that represents the analyzed area (A). Magnification: 3000x. ... 113 Figure 3.40 – SEM-EDS map of MS-Y2M sample (B) and respective micrograph that represents the analyzed area (A). Magnification: 3000x. ... 113 Figure 3.41 – Cracked sintered CERMET of 2Y+M (PA). ... 118
ix
Figure 3.43 – Simulation for several critical factor of the thermal conductivity of the CERMET (k) versus the volume fraction of metal 98. ... 120
Figure 3.44 – VSM magnetization curve of A: 2Y powder, 2Y and 2Y (PA) sintered pellets; B: metal alloy (M). ... 121 Figure 3.45 – VSM magnetization curves for CERMET samples: powders and sintered compacts. On the bottom of the image, there are two amplified graphs of the central area of the main graph in order to evaluate the coercivity of the samples. ... 122 Figure 3.46 – VSM magnetization curve of 2Y+M (PA) sample at room temperature and 5K, on the left. Magnification of the M-H loop on the right. ... 123
xi
Table 1.1 – Typical properties of commercial YSZ 11 ... 5
Table 1.2 – The properties of 2YSZ Innovnano powders. Adapted from 33. ... 13
Table 2.1 – Sample designation and description of all studied powders. ... 45 Table 2.2 – Experimental conditions to perform CPS analysis ... 58 Table 2.3– Powder shape as respective shape facto, f, and packing fraction. 82 ... 59
Table 2.4 – Gridding and polishing steps to prepare sintered pellets to flexural and hardness tests. ... 66 Table 3.1 – Identified XRD phases, calculated wt.% of each crystalline phase and respective crystallite size, strain and unit cell volume for YSZ powders (2Y (DET.), 2Y and 2Y (PA)), and YSZ CERMET powders under study in this work. The parameters were calculated with PowderCell software. ... 74 Table 3.2 – XRF results for all YSZ powders (2Y (DET.), 2Y and 2Y (PA)) and YSZ CERMET powders under study in this work. The first lines of the table correspond the major oxides and the last one to the minor oxides ... 76 Table 3.3 – Morphologic characterization of YSZ and YSZ CERMET powders: d50 measured by CPS
and Malvern equipment, surface area measured by BET (S), theoretical density of the solid material (ds) and GBET is the grain size calculated from S and ds, according to eq. 6. ... 78 Table 3.4 – Density and flowability of powders and green density of isostatically pressed compacts. ... 80 Table 3.5 – Dilatometric analysis complementary table: Green density of compacts and respective relative density, total shrinkage, thermogravimetric weight losses and final density of the powders are presented. ... 81 Table 3.6 – BET surface area, grain size and crystallographic features of the metal alloy. ... 84 Table 3.7 - Identified XRD phases, calculated wt.% of each crystalline phase and respective crystallite size, strain and unit cell volume for YSZ sintered bodies: 2Y and 2Y (PA). The parameters were calculated with PowderCell software. ... 101 Table 3.8 – Density and after-sintered characteristics of ceramic sintered bodies. Green isostatic density; weight losses (green-calcined and calcined-sintered), geometric final density, respective densification calculated based on the theoretical density (calculated based on the XRD phases wt.%), and grain size measurements. ... 102
xii
MS-Y2M. The parameters were measured with PowderCell software. ... 105 Table 3.10 – Density and after-sintered characteristics of EDS and MS-CERMETs sintered bodies. Green isostatic density; weight losses (green-calcined and calcined-sintered), geometric final density, respective densification calculated based on the theoretical density (calculated based on the XRD phases wt.%), and grain size measurements. ... 107 Table 3.11 – Flexural strength testing data, tetragonal phase wt.%, densification % and grain size of all the sintered bodies. ... 115 Table 3.12 – Vickers hardness (HV) and fracture toughness (For HV1, HV10 and/or HV30 indentations) of the sintered pellets. ... 117 Table 3.13 – Room temperature thermal conductivity and permittivity (at 10 kHz) for sintered ceramics (2Y and 2Y (PA) and CERMETs (2Y+M (PA), MS-1M and MS-2M). ... 120 Table 3.14 – Magnetic features of powders and wt.% of metal alloy calculated based in equation 24. ... 124
xiii AC Alternating current
CERMET Composite material of ceramic and metal BET Brunauner, Emmet and Teller
CPS Disc centrifuge particle sizer CIP Cold Isostatic Pressing
d50 50% of the particles have size smaller than the respective value
DSC Differential scanning calorimetry DTA Differential thermal analysis EDS Emulsion Detonation Synthesis
EXAFS Extended X-ray absorption microscopy
FM Ferromagnetic
FT-IR Fourier Transform Infrared Spectroscopy GB’s Grain Boundaries
HIP Hot Isostatic Pressing
HRTEM High Resolution Transmission Electron Microscopy ICPS Inductively Coupled Plasma Spectroscopy
MMCs Metal Matrix Composites
MS Mechanosynthesis
m-ZrO2 Monoclinic ZrO2
n-YSZ Nanocrystalline Yttria Stabilized Zirconia PSZ Partially Stabilized Zirconia
R.T. Room temperature
SEM Scanning Electron Microscopy SEM-EDS Energy-dispersive X-ray spectroscopy SOFCs Solid Oxide Fuel Cells
SPM Superparamagnetic
SQUID Superconducting Quantum Interference Device SSA Specific Surface Area
STD Standard deviation
T Temperature (K)
TEM Transmission Electron Microscopy TG Gravimetric thermal analysis TZP Tetragonal Zirconia Polycrystals
x-TZP x mol Yttria stabilized Tetragonal Zirconia Polycrystals t-ZrO2 Tetragonal ZrO2
VSM Vibration sample magnetometer W/O Water-in-oil
xiv Y-TZP Yttria Stabilized Tetragonal Zirconia
xv
θ Bragg’s angle
A Area
Aind Area of indentation of HV test
C Capacity
CBET BET constant
D Diameter
dA Archimedes density dbulk Bulk density
dG Geometrical density
dS Theoreticla density of the solid dtapp Tapped density
dth Theoretical density
E Young modulus
f frequency
F Applied load in HV teste fc percolation threshold fcr Critical factor
GBET Particle size calculated by BET surface area GBET BET calculated grain size
H Hardness
Hc Coercivity HRA Macro hardness HV Vickers hardness HV1 Vickers hardness 1 HV10 Vickers hardness 10 HV30 Vickers hardness 30 k Thermal conductivity KIC Fracture toughness KW Wear coefficient L Thickness m Weight
mbeaker Beaker Weight
mcup Cup Weight
mexp Sample+cup Weight mi Initial Weight
Mr Magnetic remanence Ms or CoM Magnetic saturation
NA Avogadro Number
xvi
R Radius
R Resistance
Rc Critical radius S BET surface area SF Particle factor shape
Sf Applied tensile stress for failure V Volume of adorbed gas
V0 One mole gas volume (at STP)
Vcup Cup volume
Vfinal Final volume
Vi Initial volume
Vm Monolayer volume
Z Impedance
α Thermal expansion coefficient γ Electrical conductivity δ Phase angle ΔG Gibbs energy ΔH Entalpy ΔS Entropy ε Permittivity ε0 Vaccum permittivity
εr Relative permeability/dielectric constant
σ Adsorbed gas area σbending Bending strength
σcompression Compression strength
σflexural Flexural strength
σy Yeld stress
3
1.1 CERMETs
Chapter 1 gives an overview on the topic of this thesis: ceramic – metal composites, normally designated as CERMETs.
This chapter presents a general definition of CERMETs, enumerates the different types of CERMETs and discusses materials performance based on the interaction of the second metallic phase within the ceramic matrix. Due to the exceptional mechanical properties of YSZ and to the fact that YSZ is currently the major product of INNOVNANO, an overview of its properties, relations with the structure, processing and application is also presented. Particular emphasis will be given to the relations between structure and properties. The phase diagram for YSZ is described and due to the current importance of nanotechnology the changes that occur in this phase diagram when dealing with nanosized powders are also presented. After, INNOVNANO company is described as a producer of YSZ nano-powders, and its proprietary EDS production method is analyzed. The main products of INNOVNANO are also described.
Finally, due to the direct relation with this thesis YSZ CERMETs are reviewed. The importance of nano-sized products, the mechanical properties of CERMETs, the magnetic response and the sintering behavior are some of the topics that will be discussed.
The name “CERMETs” was introduced after the II World War. It is composed of the syllables “cer” from ceramics and “met” from metals. Originally this expression was supposed to describe materials which combine the favorable material properties of ceramics (hardness and wear resistance) with those of metals (toughness, especially) 1.
There are several definitions of CERMETs in use. According to R.M. German 2 a CERMET is a
particulate composite consisting of ceramic particles bonded with a metal matrix. Kolaska and Ettmayer 3 define CERMETs shortly as sintered hard metals based on TiC, Ti(C,N) but with the
exclusion of WC–Co hard metals. Finally W. Lengauer 4 gives a more precise definition: CERMETs
are based on Ti(C,N) and exhibit, therefore, a purely cubic face centered hard material phase. They exhibit high wear resistance at high cutting rates if compared to conventional WC–Co hard metals. They also show high lifetimes and a good surface quality of machined materials. Typically, the hard material particles of CERMETs show a typical core–rim structure which is formed by the varying chemical stability of its components as well as by interaction between the molten binder metal and the hard phases during liquid phase sintering 1.
Non-conductive ceramic matrix, with high hardness and wear resistance, have been produced with dispersed electrically conductive secondary phases. Usually, the dispersed
4
conductive phase (e.g. TiC, TiN, TiCN, WC or TiB2) raises the electrical conductivity of the composites
above the threshold of 1 S/m. This property enables the products to be produced by electric discharge machining, for instance 5.
The design of new superhard materials (with hardness H > 40 GPa) has been a current challenge to scientists and engineers. The materials based on nanoparticles are one of the studied areas. The first propose was for the substitution of diamond particles in high resistance tools. The diamond tools cannot be used for machining steel, because it reacts with iron and silicon. Previous experimental results of hardness of single-phase nanostructured metals or metallic superlattices clearly indicate that hardness increases with decreasing grain size (between 20 and 100 nm) up to 5-7 times following a d-1/2 dependence known as the Hall-Petch effect. However, this trend inverts
for particle sizes below 20 nm (inverse Hall-Petch effect) for which hardness decreases due to a grain sliding process along particle boundaries. The origin of superhardness in these composites is attributed to: 1. the suppression of dislocations due to the small crystal size of nanoparticles; 2. the supermodulus effect in the nanocrystal core due to the compressive stress of the noncrystalline shell; 3. a strong interaction in the interface between different components 6.
Over the last decades, it is increasingly being recognized that new applications for materials require functions and properties that are not achievable by single phase materials. Combining dissimilar materials for these new applications creates interfaces whose properties and processing need to be understood to bridge the gap between the composite material microstructure and the end-product 7.
Different types of composites have been emerging, such as YSZ CERMETs. In this work, YSZ with metal addition are designated as CERMETs, despite the most usual designation as composites. YSZ is consider as the ceramic phase in those CERMETs and the metal phases can be based in nickel, titanium and cobalt metals or based metallic alloys, among others.
5
1.2 Yttria-stabilized Zirconia (YSZ)
Zirconia-based ceramics, as yttria stabilized zirconia (YSZ) and ceria stabilized zirconia present chemical stability, dielectric characteristics, high coefficient of thermal expansion, and crystal structural and lattice constant similar to that of silicon. Compositions within this system find such widespread applications as thermal barrier coatings in gas turbines, electrolytes in fuel cells, and high-temperature crucibles 8–10. Table 1.1 present a range of typical properties of commercial YSZ.
Table 1.1 – Typical properties of commercial YSZ 11
Physical properties
Density 5.85 – 6.10 g/cm3
Mechanical properties
Elasticity modulus (E) 200 – 210 GPa
Compressive strength 2200 – 2500 MPa
Electrical properties (at R.T.)
Electrical resistivity >1×1012 ohm.cm
Dielectric constant 29
Dielectric strength 9.00 – 19.0 kV/mm
Dissipation factor 0.001 – 0.002
Thermal properties
Coefficient of thermal expansion (linear) 10.3 – 11.0 µm/m.°C
Thermal conductivity 2.20 – 2.50 W/m.°C
These features predestine YSZ for a variety of applications: refractory ceramics, ceramic glazes, thermal-barrier coatings (TBCs), electroceramics, insulators, solid oxide fuel cells, oxygen sensors and abrasives, grinding media and machining tools. The advantageous characteristics of ZrO2-based materials become improved when these materials are produced from nanostructured
powders 9,10.
There are several chemical methods to produce YSZ: spray pyrolysis, combustion synthesis, hydrothermal synthesis, sol-gel synthesis, polymeric complexing methods, and EDS (INOOVNANO’s patent) 12,13.
1.2.1. Phase characterization and transformation toughening
Zirconia-based ceramics have been demonstrated to be one of the strongest and toughest oxide yet produced (as exemplified in Table 1.1 – Mechanical properties). The reason why zirconia-based products have these features will be explained later on.
6
Pure ZrO2 can exist in three crystallographic forms: cubic (c-ZrO2), tetragonal (c-ZrO2) and
monoclinic (m-ZrO2). All of these phases are variants of the cubic fluorite structure. At room
temperature, monoclinic phase is stable. The cubic to tetragonal transformation occurs at about 2340 °C and the tetragonal to the monoclinic one at about 1170 °C 14,15. Figure 1.1 shows a
representation of the crystallographic structures of zirconia.
Figure 1.1 – Representation of crystallographic structures of Zirconia 16.
The addition of soluble oxides in zirconia (MgO, CaO, Sc2O3, Y2O3 or CeO2) decreases the
tetragonal to monoclinic (t ↔ m) and cubic to tetragonal (c ↔ t) transformation temperatures. These additions are therefore said to stabilize the high temperature phases. Doping ZrO2 with such
oxides can suppress this phase transformation, and a metastable cubic fluorite solid solution can be obtained at room temperature (when more than ≈8 mol% of the yttria stabilizer is added). Such a system is referred to as the fully yttria-stabilized zirconia (YSZ). Doping with lower valence oxides (as the previously stated) can introduce oxygen vacancies into zirconia crystal lattice. When doping with a smaller amount than needed to produce fully stabilized zirconia, partially stabilized zirconia (PSZ) or tetragonal zirconia polycrystals (TZP) can be obtained. PSZ consists of fine (< 1 µm) inclusions of tetragonal zirconia in a cubic matrix. PSZ is obtained by sintering in the tetragonal + cubic two-phase field at relatively high temperature. These oxides have very high fracture toughness due to the “transformation toughening”. Yttria TZP typically contains 1.5–3 mol% Y2O3,
where some amount of cubic zirconia will be present in the microstructure if the amount of yttria exceeds about 4 mol% 8,15,17.
If the yttria’s amount is about 3 mol%, YSZ structure becomes tetragonal (t-ZrO2) after
heating above 1000 °C and the t-phase is metastable below 1000 °C where it coexists with the monoclinic one (partially stabilized YSZ). In the pure isolated oxides (yttria and zirconia), none of
7
the high temperature phases (tetragonal and cubic) can be retained by quenching to room temperature 10,14,18.
Because of stoichiometry violation, an addition of the Y2O3stabilizer introduces a large
amount of O2- vacancies or vacancies aggregate with a metal atom in the ZrO
2host lattice. The O
2-vacancies existence in YSZ leads to the ionic oxygen conductivity of the material. Further types of open-volume defects become of importance in the YSZ nanomaterials due to a significant volume fraction occupied by grain boundaries (GB’s): GB-associated vacancy-like misfit defects, vacancy clusters at GB’s intersections (triple points), voids and pores 10. The cubic polymorph has the ability
to conduct oxygen ions due to the high oxygen vacancy concentration, which increases when temperature rises. This characteristic of the cubic polymorph is the reason for its applicability in solid oxide fuel cells (SOFCs) and oxygen sensors 19.
The tetragonal to monoclinic transformation of zirconia has a great technological significance, due to its reversible and diffusionless (known as martensitic) features, with a hysteresis of 100 °C and an expansion of 4 to 5 vol.% on cooling. Because of this fact, any pure zirconia sintered block will suffer from multicracking and spontaneous failure on cooling. This detrimental mechanical instability is suppressed by PSZ and TZP 7.
The martensitic transition ability to harness the volume expansion of the structure leads to the interesting properties of high strength and toughness displayed by many zirconia products 15.
When a crack develops on zirconia surface containing metastable t-ZrO2, it is subjected to a
macroscopic tensile stress. This tensile stress concentration at the crack tip causes the transformation of metastable t-ZrO2 to the monoclinic crystalline phase. The consequent volume
increase of the crystals, constrained by the surrounding ones, results in a favorable compressive stress which acts on the surfaces of the crack, and thus hinders its propagation. Such a mechanism has been defined as “transformation toughening” or “phase transformation toughening” 7,20.
The large volume and shape deformations, which occur through the martensitic transformation, set up large strains in the structure. These strains cannot be relieved by diffusion. Instead, they are accommodated by elastic or plastic deformation of the surrounding matrix. Different models for the nucleation controlled martensitic transformation in zirconia have been developed. Although it is agreed that the martensitic reaction is nucleation controlled 15.
In Y-TZP's (Yttria stabilized tetragonal zirconia polycrystals) nucleation has been observed to occur at grain corners and has been shown to be easier in faceted intergranular grains compared to spherical intragranular grains of equivalent dimensions. Such behavior supports the theory of a “stress assisted” transformation mechanism. The stress, which is seen within the material being
8
subject to a propagating crack, is often sufficient to initiate the martensitic transformation. The subsequent volume expansion is in effect a “crack stopping” force and leads to the high values of strength and toughness recorded for these materials 8,15. Gupta et al. discovered that very fine
grained (0.2–1.0 µm) single-phase tetragonal zirconia, TZP, exhibits similar properties 21.
Nowadays YSZ is of great importance in many applications, like electrolytes for solid oxide fuel cells (SOFCs) and oxygen sensors, refractory materials for high temperature furnaces as well as protective coatings for metals 8.
1.2.2. Phase diagrams
The amount of alloying oxide required to produce stabilized zirconia is determined from the phase diagram. The phase diagram for the zirconia-yttria system is shown in Figure 1.2. Consider, for instance, in the phase diagram, compositions containing 6 mol% Y2O3 equilibrated at various
temperatures and then quenched to room temperature sufficiently rapidly to prevent cation diffusion. Above about 2300 °C such a material will be single phase with the fluorite structure, and on quenching will undergo a diffusion less transformation to a multiply twinned tetragonal phase. At 2000 °C the equilibrium is a two-phase mixture of tetragonal solid solution containing about 2 mol% Y2O3 and fluorite solid solution containing about 8 mol% Y2O3; when it is quenched the t-phase
transforms to monoclinic and the fluorite phase to tetragonal 14.
For the same composition, at 1400 °C, the equilibrium is again two phase with a tetragonal phase containing about 4 mol% Y2O3 and a fluorite phase containing about 14 mol% Y2O3; on
quenching the tetragonal phase transforms to monoclinic but the fluorite phase now contains sufficient yttria to retain that structure at room temperature and does not undergo any transformation 14.
Bulk materials behave differently of nanostructured materials, as is well known. In the specific case of nanocrystalline YSZ (n-YSZ) particles with small grain size and high specific surface help to reduce the phase transformation response time. The high surface area also causes high catalytic activity. n-YSZ properties are attributed to the large fraction of atoms within the interface region. It was shown that in systems with nanosized particles, the effect of surface area becomes significant and affects the Gibbs energy of each phase. Furthermore, the stability regions in an n-YSZ system can be significantly different from those in bulk n-YSZ. To gain a better insight into the behavior and further improve of performance of YSZ system, it is essential to understand the thermodynamic properties of these materials 8,19,22.
9
Figure 1.2 – Phase diagram of Zirconia-Yttria system. The figure shows the zirconia rich zone (mol% of ZrO2
> mol% of Y2O3) 15.
Hence, in order to develop a specific phase diagram for n-YSZ, several studies 19,22–24 have
been published. The total surface energy of particles depends on surface area and specific surface energy. As so, to have a Gibbs energy definition for a nanostructured system, the specific surface energy has to be measured 19.
Some aspects that are important to consider when designing a phase diagram for n-YSZ (this will be presented based on the Mohammad Asadikiya et al. study):
1. Calculation of Gibbs energy for bulk materials
For bulk materials, the Gibbs energy is given by equation 1, in which H is the enthalpy, T is temperature and S the entropy.
𝛥𝐺𝑏𝑢𝑙𝑘 = 𝛥𝐻 − 𝑇𝛥𝑆 Eq. 1
The model used for bulk m-ZrO2 and t-ZrO2 is (Y3+,Zr4+)1(O2-,Va)2. In this model, the first
sublattice is occupied by Y3+ and Zr4+ ions and the second one is occupied by an O2- ion and a vacancy.
The model used for c-ZrO2 is (Y,Y3+,Zr,Zr4+)1(O2-,Va)2. Since the yttria concentration in m-ZrO2 is
extremely low (as suggested by the phase diagram in Figure 1.2), this phase can be treated as an ideal solution, and the interaction parameter for m-ZrO2 phase was considered to be zero 19.
10
2. T-zero temperature method to determine phase boundaries
The phase transition in YSZ system are observed with a considerable delay, since the kinetics is very slow. In order to determine the phase boundaries under such conditions, Kaufman and Cohen 25 suggested a T-zero temperature approach (partial equilibrium method). Based on this
method, the starting transition temperature of t-ZrO2 m-ZrO2 during cooling and the starting
transition temperature of m-ZrO2 t-ZrO2 to on heating are captured. Based on these two
transition temperatures, T0 is calculated as an average temperature of previews ones. In other
words, T0 is a temperature at which the Gibbs energies of two adjacent phases are equal in a
determined composition. This T0 temperature is located in the two-phase region and it is a
theoretical limit for a diffusionless transformation 19.
Mohammad Asadikiya et al. developed the n-YSZ phase diagram at room temperature, using the T-zero method and shown that for n-YSZ there is a phase diagram area where the stabilization of t-phase is possible almost until room temperature - Figure 1.3.
When dealing with systems that have extremely slow kinetics, as is the case of YSZ, it is highly possible that heated samples will not reach the equilibrium. Hence, the measured enthalpy of such samples will be different to that of the same sample in its final equilibrium 19.
Figure 1.3 - Partial phase diagram of bulk YSZ, with indication of T0 and T’0 lines, being T0 the M/T T-zero
temperature line and T’0 is the T/C T-zero temperature line. Adapted from 19.
Mohammed Asadikiya et al. study allowed us to conclude that the phase stabilization in YSZ system is a complex matter. Moreover, the size of YSZ particles, the amount of yttria and temperature of the system are fundamental parameters. When dealing with nanosized YSZ, the stabilization of high temperature phases (tetragonal and cubic) is facilitated, i.e., it occurs at low temperature, because the grain boundaries phenomena overcame the equilibrium Gibbs energy.
11
1.2.3. YSZ Innovnano products
1.2.3.1. Innovnano, the Company
INNOVNANO emerges as a manufacturer of nanostructured zirconia and zirconate ceramic powders since 2009. The synthesis technology is unique, and therefore, patented. This unique production method, Emulsion Detonation Synthesis (EDS), guarantees a small grain size (maintenance of the intrinsic nanostructure) and chemical homogeneity. The high performance produced powders support high-tech industries as biomaterials, energy materials, electronics and sensors and thermal barrier coatings 26–28.
1.2.3.2. EDS technology
Emulsion Detonation Synthesis (EDS) method was developed and patented by INNOVNANO
28. Dynamic shock induces chemical reactions, that is known for a fact. The EDS process is based on
the detonation of two water-in-oil (W/O) emulsions, one is the initiator and the other is the secondary one. This process occurs at extremely high pressures and temperatures (> 10000 bar, from 500 to 3000 ˚C) in one single step. This kind of nanostructured powder production allows a high purity (> 99.9%) and an industrial level of production 29.
The shock wave induced in the EDS reaction leads to high pressure reactions to occur in micro-seconds, that by other process would take hours or days, enabling the synthesis of large quantities of high pressure synthesized products 29.
The reaction of synthesis takes place by combining the high temperature, high dynamic pressures and quenching. This potentiates the production of already known materials with improved properties and also the synthesis of compounds hard to obtained by common synthesis techniques, both at industrial scale and with massive costs 29.
The EDS process tries to take advantage of the relation between the diameters of the first stable liquid particles to be formed during gaseous phase reaction and the saturation degree of metal-oxide (ZrO2, for instance). Gaseous phase comprises three stages in the synthesis of
nanoparticles: 1. production of the compound in the vapor phase, 2. condensation in the form of nanoparticles and 3. control and preservation of the dispersed nanocrystalline state 29.
There are several reasons that make water in oil (W/O) emulsions particularly suitable for the powder synthesis by EDS: 1. Complete chemical reactions during detonation, assured by the high homogeneity grade of all the components; 2. Flexibility in terms of possible precursors and
12
components that allows controlling the purity and final properties of the powders; 3. Stability and safety of the emulsion, due to the high water content 28,29.
These unique synthesis technologies sets INNOVNANO apart from other manufactures. EDS allows significant flexibility in the choice of precursors and components. Great control over the purity, chemical composition, structure, morphology and final properties of the powders are some of the parameters that can be easily managed 28,30.
The specific properties of YSZ nanostructured powders prepared by EDS are: 31,32
High chemical homogeneity with uniform yttria distribution High density
Uniform grain sizes with high specific-surface area Low sintering temperatures
Enhanced physical and chemical properties
As is commonly known, ceramics powders need very high temperatures to sinter (usually above 1000°C). Therefore, the high temperatures induce undesirable grain size growth. The EDS produced powders allows the densification at lower temperatures (currently 100-150°C lower than conventional powders) due to its nanometric size 31.
13
1.2.3.3. 2YSZ
Innovnano’s 2 mol% YSZ is a product that, depending on the powder treatment, can be provided as a ready-to-press binderless powder or a suspension. The suspension can have up to 40 wt.% of solid content. These powders produce high performance ceramics with high fracture toughness and bending strength that complement each other 33.
Table 1.2 shows the cataloged chemical and physical properties of Innovnano’s 2YSZ product.
Table 1.2 – The properties of 2YSZ Innovnano powders. Adapted from 33.
Chemical analysis (wt.%)
ZrO2 + HfO2 + Y2O3 + Al2O3 >99.9 HfO2 <3.0 Y2O3 3.5 ± 0.3 Al2O3 0.4 ± 0.1 SiO2 <0.015 Fe2O3 <0.02 Na2O <0.005Other oxides (CuO, ZnO, MgO, CaO) <0.07
Physical analysis
Crystallite size (nm) 20 Primary particle d50 (nm) 50 Powder d50 (nm) <250 Granule d50 (µm) 60 Specific-surface area (m2/g) 25 ± 3Mechanical properties after sintering
Grain size (µm) < 250
Density (g/cm3) up to 6.07
Hardness (HV10) up to 1350
Bending strength (MPa) up to 1800
Fracture toughness (HV10) (MPa.m0.5) up to 15
14
1.2.3.4. 3YSZ
The 3 mol% Yttria-stabilized Zirconia (3YSZ) nanostructured powder is one of the most produced powders in Innovnano. The ceramic products made from this powders are designed for structural applications, offering high strength, enhanced fracture resistance and excellent tribological performance to endure physically demanding applications 32.
The nanostructure of this powder provides it with excellent sinterability, enabling effective processing at 50 to 75 °C lower than conventional micropowders, saving energy and retaining the powder’s nanostructure for improved ceramic properties 32.
The 3YSZ powders can be sold as spay-dried granulated powder, with and without binder (ready-to-press), as slurries and as suspensions. The atomized powders are especially suitable for hot isostatic pressing (HIP), which can improve material features, resulting in a highly dense material with reduced porosity 32.
The properties of the respective sintered part lead to its cataloging as a structural ceramic. The applications are hard-wearing components, coatings, tiles and devices in multiple industries, including the biomedical sector, industrial foundries and steel plants, mining, chemical industries and pharmaceutical manufacturing. Some particular application examples include pump linings, valve components, nozzles, tank linings or ceramic sleeves and cutting tools 32.
15
1.3. Interaction between metal and ceramic matrix
In order to explain the behavior of composite materials, in which the constituent phases have different properties, percolation theory emerges.
Percolation theory is a well reported mathematic theory and largely studied subject. It describes the geometry and behavior of connected clusters in a random space, modeling the richness of the various interconnections that may be present in a random grid. Percolation theory explains why certain properties of multiphase materials undergo a dramatic change when one of the phases reaches a critical concentration, fc, instead of following the linear rule of mixtures. This theory tries to explain physical properties such as electrical resistivity, optical properties, thermal conductivity, among others 7.
To understand the range of new properties emerged in composites materials, percolation theory is an important tool, especially in those materials whose phases (matrix and particles) present very different values of a given property, such as CERMETs. Therefore, the percolation theory is a powerful tool for designing new materials in the range of the percolation threshold (fc)
7. In the specific case of this work, the percolation theory was not applied to select the amount of
second phase because this has been previously selected by INNOVNANO.
The theory says that the infinite metal cluster) formed at the percolation threshold can be used to release internal stresses induced into the bulk of the material. This happens because of the thermal expansion mismatch between different phases or because an enantiomorphic phase transformation, such as α ↔ β quartz, martensitic transformations, among others, occurs. The last example is the case of the tetragonal to monoclinic (t ↔ m) ZrO2 transformation. At the percolation
threshold and above there is an infinite cluster that ensures the conductivity in the system 7,34.
Academically, a ceramic/metal interface is known as a contact between two classes of materials that usually have very different properties from each of the materials due to their different bonding characteristics. The difference in material properties between the metal and the ceramic induces stress singularities at the interface. The stress singularity combined with the thermal residual stress can degrade the strength of ceramic/metal joints. As so, microstructural development at ceramic/metal interfaces plays a critical role in all of these processes. The interfacial morphology can determine the performance characteristics of dissimilar material joints, such as metal–matrix composites, ceramic–matrix composites, electronic packages, glass-to-metal seals, bioglass–metal coatings, metal to dental ceramics joining, etc 7.
Different systems behave differently. For instance, Al2O3/Ni composites were proved to have
16
this mechanism is fully operative only when the metallic inclusions are strongly bonded to the brittle matrix. If the ductile phase inclusions are weakly bonded to the ceramic matrix, the cracks will propagate along the ceramic/metal interface, and the contribution of the ductile particle to improve the toughness of the final dense composites will be negligible. This weak bonding is observed in composites of 3 mol% yttria tetragonal partially stabilized zirconia (3Y-TZP) with nickel. The addition of Ni particles to 3Y-TZP matrix does not increase the toughness of the composites. The electronic structure of the interfaces must be the reason for the weak ZrO2/Ni interface
17
1.4. TiC CERMETs
As said before, TiC were the first referred and defined CERMETs. In recent years, the interest in TiC/TiCN-based materials has been growing because of their good high-temperature hardness, superior thermal conductivity, excellent creep resistance, relatively low friction coefficient, perfect high-temperature oxidation resistance and chemical stability. 35
This family of structural and wear-resistance materials can compete with WC-Co hard metals in several applications, in particular machining of steel, and they are superior in surface finishing operations. Industrially, these materials have another advantage in comparison with WC-Co hardmetals: they are less expensive. As so, Ti(C,N)-based CERMETs have been widely used as high-speed cutting tool materials for semi-finish and finish machining of carbon steel, stainless steel and alloy steel. They can provide a better geometry accuracy control and surface quality35,36.
Usually, Ti(C,N)-based CERMETs are composed of two phases, as previously mentioned: one is a hard ceramic phase, carbonitride particles that can provide high hardness, and the other is a metallic binder phase, for instance nickel or cobalt which gives contributions to the strength and toughness of the material. The microstructure of Ti(C, N)-based CERMETs is typically characterized by carbonitride particles, exhibiting a ‘‘core-rim” structure, bonded with a metallic phase. It is well known that different final mechanical properties of these composites depend on different chemical composition and microstructure of “core-rim-binder” phase. The microstructure and performance can be engineered by varying properly the chemical composition, such as Co, Mo, Cr, Mo2C, WC,
TaC, NbC ,Cr3C2, AlN, C among others 35.
Submicron or nanometric carbide powders have been developed as cutting tools with ultra-fine microstructures. The use of small size powders in hard metals and CERMETs improves greatly their mechanical properties. However, the fracture toughness is usually inversely proportional to the hardness and grain size, unless the grain size is extremely fine or at a nanoscale. Recently, ultra-fine TiCN-based CERMETs with completely dense bodies showed a high hardness (Vickers Hardness (HV) with 30 N load, HV30 = 1800 MPa) but with a very limited toughness (7–9 MPa.m1/2). Although,
this toughness can be improved by using high binder content, increased grain size of the ceramic phase or using (Ti,W)C solid solution powders instead of binary powders. However, the hardness will be weakened, leading to dissatisfied wear resistance 36.
TiC powders are industrially prepared through TiO2 carbothermal reduction at high
temperatures (1700 °C – 2100 °C). Subsequently, TiC powders and other raw materials are uniformly mixed in certain proportions to produce the TiC-metal. Finally, the TiC CERMET can be prepared by vacuum sintering at high temperature 36.
18
1.5. WC CERMETs
Another example of a composite family, sometimes called CERMET composites and largely utilized in the industry, as a substitute of the high resistance steel working pieces, is the tungsten carbide/cobalt system (WC/Co). This chapter gives an overview on this group of materials. Some composites of WC are presented, as WC with Co/Ni, WC-Co with YSZ and Ni/W matrix with zirconia.
These cemented tungsten carbides consist of the hard carbide phase WC embedded in a ductile metallic matrix, usually referred to as binder phase. For cemented tungsten carbides, the great properties of the components are superimposed; the carbide phase WC provides hardness and wear resistance while the ductile binder (transition metal as Co, Fe or Ni) contributes to toughness and strength, in accordance with what occur with TiC CERMETs. Due to the excellent combination of hardness and toughness, cemented tungsten carbides are often referred to as “hard metals” and are used widely in cutting, rock drilling and molding. Hard metal has been industrially produced since the 1920s and nowadays represents more than 40% of the cutting tool market 37,38.
1.5.1. WC matrix with (Co/Ni)
Despite of being produced for several decades now, during the last years, the deployment of cobalt natural resources and the increasing demands on material performance had led researches to the optimization of new binder compositions and the development of specific coatings to improve the useful properties 38.
C.M. Fernandes et al. have presented a method of preparing composite powders of WC and low Ni/Fe/Cr binder content (4–7 wt.%) composites, consisting of sputtered metallic binder onto WC particles and Figure 1.4 presents the microstructures of the composites prepared from those powders and from conventionally milled ones. The coated powders resulting from this technique show a very high uniformity of binder distribution associated with a nanocrystalline structure. The surface properties of the particles were changed, increasing the powder’s flowability, pressing behavior and sinterability in such a way that easier powder processing could be adopted 38. In C.M.
Fernandes’s study, depth-sensing indentation was used to measure the mechanical properties in compressive strength, with a small size sample. The characteristics obtained for a standard WC–Co sample using this method were very close to the published results of macroscopic characterizations for the determined values of hardness, H, Young’s modulus, E, and yield stress, σy. They concluded that the lower values of E of the sintered WC–Ni/Fe/Cr compared to those found in similar sintered conventional powders cannot be attributed to differences in grain size as for H and σy. The high binder uniformity and the nanometer-sized coating achieved by the sputter-deposition process is
19
not the principal cause for the lowest values of E. They proved that the decrease in E may have its origin in Ni diffusion into WC powders during the sputtering, which is enhanced during the sintering process. However, a preferential orientation could also have a role in E values, contributing to their decrease. It was suggested that lower amounts of ductile binder could be needed to obtain convenient ductile properties in the composites prepared from sputtered powders, than when a conventional mixing process is used 38.
Figure 1.4 – Optical micrographs of a composite prepared from WC powders coated with iron rich binders (a) and a conventionally prepared WC composite with the same binders (b). The surrounding area shows the heterogeneity of binder distribution. 38.
Wei Su et al. recently prepared and studied WC with 6 mol% (Co, Ni) composite powders with different Co/Ni ratios. These composites were fabricated through hydrogen reduction of WC– (Co,Ni)C2O4.2H2O precursor. They have concluded that the liquid phase temperature and corrosion
resistance of WC–6(Co, Ni) cemented carbides increase with the increase of Ni content. When 20 wt.% Co is substituted by Ni, the average WC grain size decreases from 1.67 μm to 1.48 μm, the hardness and the transverse rupture strength increases from 2182 MPa to 2276 MPa, respectively, while the corrosion rate decreases and the magnetic coercivity, Hc, is increased 37. The coercivity,
or coercive force, Hc, is a measure of the ability of a ferromagnetic material to withstand an external magnetic field without being demagnetized 9. Although the addition of Ni is supposed to
facilitate the solution-precipitation of WC particles and promote the growth of WC grains, the liquid phase temperature of WC–6(Co,Ni) increases with increasing Ni content. Co substituted by Ni should have the positive and negative effects on the growth of WC grains 37.
As shown in Figure 1.5, the magnetic saturation (point when the increasing of external magnetic field cannot increase the magnetization of the material – Ms or CoM) of WC–6(Co, Ni) sintered alloys is affected together by Ni and W contents in Co, and the variation of the magnetic saturation of WC–6(Co,Ni) sintered alloys is a convex parabola. Since the magnetic saturation quantities of pure cobalt and nickel are 202 μTm3/kg and 68 μTm3/kg, respectively, the addition of
20
dissolved in the binder can further reduce the magnetic saturation of WC–6Co sintered alloys 37.
For a constant binder phase, coercive force, Hc is supposed to decrease with increasing average WC grain size of sintered alloys. In addition, coercive force increases with the decrease of Ni content and the increase of W content in Co. It can be seen from Figure 1.5 that the coercive force, or magnetic coercivity, of WC–6(Co, Ni) sintered alloy reaches the highest value of 12.4 kA/m when 20 wt.% Co is substituted by Ni 37,39.
Figure 1.5 – Effects of gradual cobalt substitution by nickel on magnetic saturation (CoM) and coercive force or coercivity (Hc) of WC/6Co alloy 37.
1.5.2. WC-Co matrix with YSZ
As described above, WC-Co CERMET is an important material the substitution of tool and die steels. According to different requirements, they should have good wearability, impact toughness, high strength and high hardness 40. Some composites of WC-Co, as WC-Co/YSZ, have shown
improvement of mechanical properties and hence become good substitutes of hard metal tools. An example of some recently developed CERMET composites is WC-20wt.%Co + 3YSZ (composite of tungsten carbide with 20 wt.% cobalt matrix with 0, 1, 2 and 3 wt.% 3YSZ) that were prepared by normal vacuum sinter processing and characterized by Lin An et al 40. Results showed
that 3YSZ spherical particles with different sizes which were uniformly distributed in Co and WC matrix phases; the bending strength and impact toughness of these WC-20wt%Co + 3YSZ composites were improved remarkably, but the hardness values had a small change 40. The macro
hardness (HRA) of the composites with 3YSZ was slightly higher than that of the specimen without 3YSZ with no distinct fluctuation of macro hardness all over the composites 40.
21
1.5.3. Ni/W matrix with ZrO
2Metal matrix composites (MMCs) have been developed with increasing interest because of the demand for advanced materials with precisely controlled properties. These class of materials have almost unlimited possibilities for material engineering, as their properties can be designed depending on the application. The properties of MMCs are mainly determined by the combination of components as well as by their interfacial characteristics, like in the CERMETs case. MMC coatings are usually designed to improve surface tribological properties, since metals are hardened by the incorporation of ceramic particles. Different types of particles with a variety of properties, e.g., oxides (Al2O3, ZrO2), carbides (SiC, WC, SiC), nitrides (Si3N4) and borides (TiB2, ZrB2), have been
commonly used to reinforce matrices of microcrystalline metals or alloys 41.
The nickel-based alloys are widely applied as composite matrices due to their superior properties. Despite that, nanostructured nickel is generally unstable, which may lead to a rapid grain growth even at low temperatures. Alloying with some metals of high melting points has been found to improve the stability of that system. Nanocrystalline Ni–W alloys are known for their stability, high hardness, high wear resistance at elevated temperatures, high melting point, low coefficient of thermal expansion, high tensile strength and high corrosion resistance in many aggressive environments 41.
Composite coatings of a nanostructured Ni–W matrix reinforced by zirconia (ZrO2)
nanoparticles are new materials. Zirconia is an extremely refractory material that offers high hardness, high wear resistance and chemical inertness 41,42. E. Beltowska-Lehman et al. synthesized
composite coatings consisting of a nanocrystalline Ni–W alloy matrix reinforced with ZrO2 particles
(average size of 50 nm) by electrochemical deposition assisted by an external field 41. All the
composite coatings were crack free, homogenous, compact and well adherent to the steel substrate. In addition, good interconnection between the phases (ceramic particles and the metallic matrix), the lack of voids and discontinuity at the interface were observed. These composite coatings exhibit a considerable enhancement in microhardness in comparison to pure Ni–W and composite Ni/ZrO2
coatings. The Ni–W/ZrO2 composites with the lowest ceramic content (about 5 wt.%) presented the
